While preparing to bring my 3D printer back to life, I came up with an interesting question…
If you control fan speed during 3D printing, wouldn’t the PID tuning give you funny results because the hotend would get hot/cold at a different speed than what is expected when the speed is at a different speed than what it was during PID tuning?
Yes, which is why its important to aim the print fan at the print, not the hotend.
@Tim_Elmore It is a hotend cooling fin fan. It does not point at the nozzle or heater block and I don’t think the breeze should hit it although I must admit that I have no shroud on it. It is point blank at the cooling fin though.
The fan control in your slicer is for a part-cooling fan.
The fan cooling your hot-end heatsink should be on any time the hot end is above room temperature. It’s usually easier to just have the fan wired to the power input, so that the fan is on whenever the printer is on.
As said previously, that fan should be on 100% speed all of the time that the hotend is hot. As a result, it doesn’t impact PID tuning.
I prefer to run mine so it comes on when the hot end reaches 50C. It saves some wear and tear on the fan. Plus it’s already built into most firmware, you just need to define which pin is activated, and your preferred “on temp”.
I use the auto on at 50 as Eric does on boards that support it. I started using in on an MM3 last year. A RAMPS won’t have a fan port available for this use unless you aren’t using the heated bed. To use this with an extruder heater, parts fan and heated bed you’ll need four PWM outputs.
My jhead is wrapped, well, only the brass pas is wrapped in ceramic wool, stopping fan air from altering the temperature. Hence, i have two fans pointing directly at the tup of the head and it works just dine and cools pla very nicely
Yes, but that’s the point of using a PID controller instead of just setting a power level.
However… You will move along the response curve to an area that might have a different slope, so the PID parameters might not be perfectly tuned if they were tuned when the fan was off, but the result is probably negligible.
If you really want to, you could implement a feed forward control in addition to PID so that the power level gets bumped up when the fan is on, but if your fan is aimed well it probably isn’t needed. There have also been experiments with bumping up the power based on the amount of plastic being melted. I think the code complications were deemed not worth it.